The green bond market is developing rapidly. By April 20201, the total global market volume had increased to USD 829 bln from USD 3.1 bln in 2012. An increasing number of issuers with different industry and geographical affiliations are entering the market, currently amounting to over 700. New companies and countries are constantly entering this market as well. Georgia, Kazakhstan, and Ukraine have already announced their first issues. Investor interest in this type of bond is also increasing due to the inclusion of environmental risks in the list of investment evaluation criteria. Despite the rapid development of the green bond market, the differences in the key characteristics of these financial instruments from conventional bonds are poorly understood.
Investors are willing to pay more for green bonds. Most research confirms “discounts” on green bond yields2 when compared to similar types of bonds. Analyzing samples from different time periods and characterized by different geographical issue distribution, researched concluded that there is a statistically significant discount on green bond yields. In general, the increased market demand for green bonds affects the growth of prices for this type of asset, which is expressed in lower levels of profitability (Fig. 2).
Figure 1. Average decline in yield to maturity after bond placement over the past two years
Source: VEB.RF
1 Data from Climate Bond Initiative (https://www.climatebonds.net/2020/05/markets-monthly-4-07052020-april-gb-issuance-bounces-back-green-stimulus-calls-grow-china).
2 In other words, the market has a negative premium to the yield level of green bonds.
Figure 2. Discounts on green bond yields compared to classic bonds
Source: ACRA
ACRA assessed 3,477 issues from January 2007 to March 2020, including data on classic and green bonds. ACRA’s model found a small but statistically significant discount of 1-2 bps3 on green bond yields compared to similar classic bonds. ACRA estimates that the lower discount values are due to different time periods of analysis (ACRA used model data updated for recent years, and as the green bond market grows and gradually moves from a stage of active growth to maturity, the discount tends to decrease). In addition, the inclusion in the sample of data on emerging markets may have had an impact on the assessment. In these markets, investor demand for green instruments is not as pronounced and the degree of information transparency on issuers and the quality of information provided by them may be lower than that of issuers from developed countries.
ACRA notes that bonds that meet a number of criteria based on the green bond principles acquire the status of “green” after passing verification. ACRA has developed its own methodology for evaluating green bonds.
3 The model was constructed using logarithm of the dependent variable in order to achieve a higher quality econometric model. In order to simplify, ACRA used an interpretation that got rid of the logarithm. However, this only gives an approximate estimate, since in this case the direct interpretation is not completely correct.
The fact that investors are willing to settle for lower returns on green bonds may indicate a few things: growing investor interest in this type of asset, awareness of the severity of environmental degradation, insufficient representation of environmental projects, insufficient volume of infrastructure investments needed to transition to a low-carbon economy, and the excess of demand over supply in this segment. The latter is often expressed in oversubscription when placing green issues, including Russian issuers4, and is due to either the internal characteristics of green bonds or their insufficient representation in the market at the moment. A lower yield may indicate that these bonds are attractive to investors even if it means lower cash flow.
4 There was oversubscription green bond issues of Joint Stock Company “Russian Railways” "and Center-invest Bank.
Green bonds compete in terms of liquidity
In order to draw more objective conclusions on the advantages of green bonds, ACRA also analyzed the degree of their liquidity. This issue is particularly relevant because it is believed that the greatest interest in green bonds is shown by investment funds and other institutional investors who purchase green securities when issued and hold them until maturity, which negatively affects the volume of trading in these securities, their liquidity, and their price in the secondary market. However, the analysis showed that green bonds have a higher degree of liquidity than classic bonds. The difference between ask and bid[2] prices for green bonds is 47.9% less than for other comparable bonds, which indicates that these bonds are more actively traded and have higher liquidity. This is undoubtedly a positive factor for potential investors.
As the market matures, discounts on green bond yields could fizzle out. Currently, the market is experiencing a supply shortage and increased demand for green bonds. However, the expansion of this market segment will inevitably lead to a balance between supply and demand. This can offset the discount effect, which is expressed in lower green bond yields. According to ACRA, the increased supply of green bonds in coming years will be primarily in those sectors with a lack of “green” investment (currently these include industrial production and financial institutions). In the industrial sector, growth will be stimulated by stricter regulatory requirements for enterprises, as well as a new course of development focused on environmentally sustainable economies in developed and developing countries. In the financial sphere, growth will be due to a significant increase in the volume of credit financing for projects in the field of ecology and sustainable development, which are steadily becoming more relevant.
Figure 3. Green bond issues by industry in 2019
Source: CBI
5 The bid-ask spread is the difference between the highest price that a buyer is willing to pay for a given asset and the lowest price that a seller is willing to accept when selling a given asset. The bid-ask spread is often used to assess the degree of liquidity of an asset in the market (the smaller the spread, the more liquidity the asset has).
In Russia, the number of green bond issuers is still small, and the corresponding market segment is developing at a fairly slow pace. However, ACRA believes that green bonds in Russia have a great potential and in coming years they will be beneficial to issuers.
According to ACRA, the most important factors that will help accelerate the development of the green bond market in Russia are: state support (special benefits, subsidies, and other incentives, such as tax incentives), reducing costs for environmentally friendly technologies, which will make them more attractive to use, and the impact of price dynamics for key energy carriers and prices for carbon emission quotas. VEB.RF will play a key role in the development of the green financial instruments sector in Russia as a factory of “green” projects aimed at modernizing the domestic economy.
Technical note
As part of the model, the following dependent variables were used to assess the difference in yield: coupon rate for bond issues, year of bond issue, volume in circulation, the minimum amount for which trades on this issue are accepted per person, Macaulay duration6, credit rating, and the US consumer price index. The OAS (option adjusted spread7 was used as a proxy for the yield spread variable, while the degree of liquidity for bond issues was assessed using the bid-ask spread. To assess if a bond is “green,” the GB binary variable was used, which takes the value of 1 if the issue is green, and 0 if it is classic.
When constructing the model, the following dependent variables were used to assess the difference in liquidity between green and comparable classic bonds: OAS, coupon rate for bond issues, year of bond issue, Macaulay duration, volume in circulation, credit rating, bond issue market (domestic or international), and coupon payment frequency. To assess if a bond is “green,” the GB binary variable was used, which takes the value of 1 if the issue is green, and 0 if it is classic.
When building models and selecting influencing factors, ACRA took into account the models of other researchers that analyze the difference in the yields of green and classic bonds (Zerbib O. D., 2018; Preclaw R., Baksli, 2015; Karpf A., Mandel A., 2017; Nanayakkara M., Colombage S., 2019; Gianfrate G., Peri M., 2019). ACRA also took into account articles on the general determinants of bond yield spreads (Garay U., González M., Rosso J., 2019; Bao J., Hou K., 2016; Lu C., Chen T., Liao H., 2010; Zhou R., Xiong Y., Liu T., Li J., 2019; Mukherjee K., 2019), as well as on the importance of assessing liquidity when analyzing bond issue yields (Febia W. Schäfera D., Stephan A., Sun C., 2018; Chen L., Lesmond D. A., Wei J., 2005; Han S., Zhou H., 2006).
6 The Macaulay duration is the weighted average term to maturity of the cash flows from a bond. Duration is used as a measure of interest rate risk.
7 The spread between the analyzed issue relative to the yield of comparable risk-free bonds, which is adjusted to account for possible pledged put or call options. When discounting cash flows from a bond issue at a risk-free rate plus OAS, the amount of discounted cash flow on the bond is equal to its current value. OAS is measured in percentage points.
